Peptide Source
Peptide source research focuses on developing methods for designing, identifying, and characterizing peptides, particularly for therapeutic applications. Current efforts leverage machine learning, employing transformer-based language models, graph neural networks, and generative models like those based on reinforcement learning, to predict peptide properties, design novel sequences with desired characteristics (e.g., antimicrobial activity, cell penetration), and improve the accuracy of peptide sequencing from limited data. These advancements are significantly impacting drug discovery and development by accelerating the identification of promising peptide candidates and optimizing their properties for improved efficacy and safety.
Papers
An Efficient Consolidation of Word Embedding and Deep Learning Techniques for Classifying Anticancer Peptides: FastText+BiLSTM
Onur Karakaya, Zeynep Hilal Kilimci
From Peptides to Nanostructures: A Euclidean Transformer for Fast and Stable Machine Learned Force Fields
J. Thorben Frank, Oliver T. Unke, Klaus-Robert Müller, Stefan Chmiela